research on two dimensional silicon photonic …...博 士 論 文 research on two dimensional...
TRANSCRIPT
博 士 論 文
Research on Two Dimensional Silicon Photonic Resonators for Label Free
Detection of Biomaterials
非標識生体物質検出のための二次元
シリコン光共振器の研究
Amrita Kumar Sana
広島大学大学院先端物質科学研究科 Graduate School of Advanced Sciences of Matter
Hiroshima University
March 2017
目 次 (Table of Contents)
1. 主論文(Main Thesis)Research on Two Dimensional Silicon Photonic Resonators for Label Free Detection of Biomaterials (非標識生体物質検出のための二次元シリコン光共振器の研究) Amrita Kumar Sana
2. 公表論文(Articles)(1) Silicon Photonic Crystal Resonators for Label Free Biosensor
A. K. Sana, K. Honzawa, Y. Amemiya and S. Yokoyama
Japanese Journal of Applied Physics 55 04EM11 (1-5) (2016).
(2) Temperature Dependence of Resonance Characteristics of Silicon
Resonators and Thermal-Stability Improvement by Differential
Operation Method
A. K. Sana, J. Maeda, Shu. Yokoyama, Y. Amemiya and S. Yokoyama
Japanese Journal of Applied Physics 56 04CC06 (1-5) (2017).
(3) High Sensitivity Double-Cavity Silicon Photonic-Crystal Resonator for
Label-Free Biosensing
A. K. Sana, Y. Amemiya and S. Yokoyama
Japanese Journal of Applied Physics 56 04CM06 (1-5) (2017).
主 論 文 (Main Thesis)
ii
Abstract Early detection as well as identification of biomolecules is very important for the
medical diagnosis of many diseases. The mostly used biomaterial detection methods are
fluorescence based detection and label-free detection. Fluorescence based detection is
quite sensitive and able to detect down to a single molecule and usually use
fluorescence tag to identify the target presence. But most complexity is the labeling
steps beyond the isolation of target analyses. In contrast, the label free detection
performs the detection of biomolecules in their original form also able to easier and
cheaper bio detection. In contrast, biosensors such as enzyme-linked immunosorbent
assay (ELISA) and surface-plasmon-resonator (SPR) which are already in commercially
available label free detection method. However, the advantages of ELISA method need
large volume of analytes, expensive and long measurements time. On the other hand the
size of SPR is quite large.
Our research focused on Si photonic crystal resonator based biosensor which is compact
and highly sensitive because the light of certain frequencies is confined in resonator and
takes interaction with biomaterial many times. Thus it has the advantages of being very
high sensitive even though the very small volume of analytes.
To confirm our device performance we first used sucrose solution (solution of sugar
and water) because of there is no need any extra binding method between device surface
and sucrose solution and it is easy to remove them from the device surface. We were
able to detect the resonance wavelength shift by changing the sucrose concentration. We
reported Q and sensitivity are 105 (Q is defined as res/FWHM, where res is the
resonance wavelength and FWHM is full width at half maximum of resonance peak),
1570 nm/RIU (Refractive Index Unit) (sensitivity is the ration of res/n, where res
is a resonance wavelength shift and n is a change in refractive index) respectively.
We also made double nanocavity type photonic crystal resonators where surrounding
air hole radius were modulated and we got the highest Q of 2 105, sensitivity of
1571 nm/RIU (s) and minimum detection limit of < 10-6 RIU (detection limit is defined
as res/QS, where Q is the quality factor and S is the sensitivity).
To detect prostate specific antigen (PSA) we used a very special feature called Si
binding protein (Si-tag) which has been developed by Professor Akio Kuroda,
iii
Department of Molecular Biotechnology, Hiroshima University. Si-tag immobilizes the
receptor anaytes on the device surface (on Si / SiO2) with a same aligned orientation.
Due to the receptor analytes, antigen-antibody is attached on the device surface in an
aligned manner and light-matter interaction improves more, as results resonant peak
shift even though the concentration is very low. We successfully detected PSA
concentration as low as 0.01 ng/mL, where the practical sensitivity is 1 ng/mL.
We describe the mechanism and solution for dominating temperature effects on
refractive index based Si optical resonator sensor such as ring resonator and photonic
crystal resonator sensors. The temperature change affects the silicon refractive index
and affects resonator mechanical shape also. As a result, it is reported that the refractive
index change is dominating whereas the mechanical deformation effect is negligible.
We also demonstrated that the differential operation is effective to suppress the
temperature effect for Si ring resonator sensor.
iv
Dedication
I dedicate this dissertation to my parents and parents-in-laws.
v
Contents Chapter 1 Introduction
1.1 Introduction………………………………………………………………..….1
1.1.1 Background of research…………………………………………………….1
1.1.2 Properties of periodic photonic crystal…………………………….………1
1.1.3 Silicon photonic crystal as biosensor……………………………………..4
1.2 Goal of this work……………………………………………………….…...4
1.3 Dissertation outline………………………………………………………….4
References……………………………………………………………………………6 Chapter 2 Si photonic crystal resonator biosensor
2.1 Finite-Difference-Time-Domain analysis of the photonic crystal resonators..7 2.2 Designing of photonic crystal resonator………………………………………7
2.2.1 Cavity type resonator………………………………………………………9 2.2.2 Defect type resonator………………………………………………………9 2.2.3 Special shape resonator (cavity connected with surrounding air holes)......10
2.2.4 Double cavity resonators with neighbor hole modulation…….……..……11 2.3 Resonance characteristics of the device………………………….………..…11 2.4 Details about Q-factor, sensitivity, figure of merit and detection limit………12
2.5 Comparison of Q value between cavity and defect type structure…………..13 References………………………………………………………………...…………..14
Chapter 3 Fabrication and measurements of photonic crystal resonators
3.1 Fabrication of Si photonic crystal resonator biosensor……………………….15 3.1.1 Reason for high Q by proposed device structures……………………17 3.2 Measurement system of Si photonic crystal resonator biosensor……………19 3.2.1 Schematic of measurement systems………………………………………19
References……………………………………………………………………………...21
Chapter 4 Detection of device performance using sucrose solution 4.1 Sucrose solution preparation ………………………………………………..22 4.2 Measurements procedures…………………………………………………….23 4.3 Details about sucrose concentration and its corresponding refractive index...24
vi
4.3.1 Measurement results ………………………………………..……….……24 4.4 Linearity check with resonant wavelength shift by the sucrose
concentration………………………………………………………………..30 References…………………………………………………………………………....32 Chapter 5 Detection of antigen-antibody reaction
5.1 Details of Si-tag, protein-G, antibody (IgG2a) and prostate specific antigen...33 5.2 Measurement procedures………………………………………………..……34 5.3 Resonance characteristics versus PSA concentrations………………….……36 5.4 Measurement results fitting with Langmuir’s curve………………………37
5.5 How could linearity improvement is possible in saturated region at higher concentration of biomaterials…………………………………….………37
References…………………………………………………………………………..38 Chapter 6 Temperature independent device for the biomaterial sensing
6.1 Temperature effects on resonance wavelength shift………………………39 6.2 Numerical analysis and simulation………………………………………..40 6.3 Schematic of device structures of differential Si ring resonators…………..45 6.4 Measured results…………………………………………………………..47
6.4.1 Mechanism of thermal change of res of Si ring and PhC resonators……..47 6.4.2 Thermal stability of differential Si ring resonator sensor………………..47 References…………………………………………………………………………..52 Chapter 7 Conclusion
7.1 Conclusion………………………………………………………………..54 7.2 Comparison among other works………………………………………….55 7.3 Impact of this research work……………………………………………..57
References…………………………………………………………………………….58
Acknowledgement…………………………………….………………59
List of publications……………………………………………………...60
1
Chapter 1 Introduction 1.1 Introduction
1.1.1 Background of research
Early detection as well as identification of biomolecules is very important for the
medical diagnosis of many diseases. The mostly used biomaterial detection methods are
fluorescence based detection and label-free detection.1) Fluorescence based detection is
quite sensitive and able to detect down to a single molecule and usually use
fluorescence tag to identify the target presence. But most complexity is the labeling
steps beyond the isolation of target analyses. In contrast, the label free detection
performs the detection of biomolecules in their original form also able to easier and
cheaper bio detection. In contrast, biosensors such as enzyme-linked immunosorbent
assay (ELISA)2) and surface-plasmon-resonator (SPR),3-6) which are already in
commercially available label free detection method. However, the advantages of ELISA
method need large volume of analytes, expensive and long measurements time. On the
other hand the size of SPR is quite large.
Our research focused on Si photonic crystal resonator based biosensor which is compact
and highly sensitive because the light of certain frequencies is confined in resonator and
takes interaction with biomaterial many times. Thus it has the advantages of being very
high sensitive even though the very small volume of analytes.
1.1.2 Properties of periodic photonic crystal
Photonic crystals7,8) are periodic nanostructures with a spatial periodicity in their
dielectric constant. Under certain conditions, photonic crystals can create a photonic
bandgap, i.e. a certain frequency of light in which propagation through the crystal is
inhibited. Light propagation in a photonic crystal is similar to the propagation of
electrons and holes in a semiconductor. The periodicity of the electronic potential in
semiconductors, which is due to the regular arrangement of atoms in a lattice, gives rise
to the electronic bandgaps, which are forbidden energy bands for electrons. Similarly,
the periodicity of the refractive index gives rise to photonic bandgaps, forbidden energy
2
bands for photons. According to the periodicity, photonic crystal are classified into three
ways i) one-dimensional (1D) ii) two dimensional (2D) and iii) three dimensional (3D)
photonic crystal.
One dimensional photonic crystal:
A one-dimensional photonic crystal (1D PhC), which is a periodic nanostructure with
a refractive index distribution along one direction. As illustrated in Fig. 1.1 the simples
structures of one dimensional photonic crystal. Because of its wavelength selective
reflection properties, can be used in a wide range of applications including Fabry-Perot
cavities, optical filter, high efficiency mirrors and distributed feedback laser. Most
recently 1D PhCs is also be using as fiber Bragg gratting (FBG), where fiber core
refractive index is changing along its axis.
Fig. 1.1. Schematic of one dimensional photonic crystal.
Two dimensional photonic crystal:
The schematic structure of two dimensional (2D) photonic crystals is shown in
Fig.1.2. In 2D PhCs, the periodicity of dielectric constant varies along the two axis
rather than one direction.
3
Fig. 1.2. Schematic of two dimensional photonic crystal.
Three dimensional photonic crystal:
In three dimensional (3D) photonic crystal lattice, the dielectric constant varies in
all three direction. 3D PhCs is the most challenging structures to fabricate.
Waveguiding and making defects in 3D structures has not progressed as like 2D
structures and need complex geometry to get 3D bandgaps. The schematic structure of
3D PhCs is shown in Fig. 1.3.
Fig. 1.3. Schematic of three dimensional photonic crystal.
4
1.1.3 Silicon photonic crystal as biosensor Photonic crystal based devices that are used for sensing applications are typically
micro-fabricated periodic structures on silicon on insulator (SOI) chip. The structures
can be designed to exhibit a two-dimensional photonic band gap (PBG) and tune by
varying the structural parameters such as lattice constant, air holes radius. In our case,
we systematically introduced cavity/defects into the perfect hexagonal lattice during the
fabrication process. The cavity was introduced by enlarging radius of the center hole
whereas defect was created by removing center hole. Such type cavities typically
exhibits small modal volumes as well as strongly localize the electric field in the cavity
or defect region is highly sensitive as changes of local refractive index. These properties
make them highly interesting for refractive index and biosensing applications. Whilst
the sensitivity of that structures are dependent on the precise cavity geometry.
1.2 Goal of this work In this PhD research work, photonic crystal resonators has been investigated to use
as a biosensor. The target of this work was to detect biomaterials i.e prostate specific
antigen (PSA) marker at a very lower concentration (below 1ng/mL). We also studied
temperature effects on resonance wavelength shift and the solution to avoid the
temperature effects by differential silicon ring resonator sensor.
1.3 Dissertation outline Chapter 1 explains the background of this thesis work. The basic of the photonic
crystals, classification and its use as biosensors are described in this section. In Chapter
2 FDTD analysis, design rules of photonic crystals resonators and the theory of
resonance wavelength shift are described. Chapter 3 describes the fabrication process
and experimental method as well. In Chapter 4 device performance was measured using
sucrose solution and prostate specific antigen. The sensitivity and quality factor of the
devices are explained in this section. Chapter 5 explains the antigen-antibody reaction
process, experimental data and the Langmuir’s fitting of the experimental results. In
Chapter 6 theory of temperature effects on resonance wavelength of silicon photonic
resonators (i.e Si ring and Photonic crystal resonators) are explained. The device
5
mechanical deformation due temperature rise is also explained. Differential ring
resonator sensor was described in this section to avoid the temperature effects. In
Chapter 7 the device performances are summarized and compare with other biosensors.
The impact of this thesis is described here in this section.
6
References [1] F. Hosseinibalam, S. Hassanzadeh, A. Ebnali-Heidari, and C. Karnutsch, Appl. Opt.
51 (2012) 568.
[2] R. M. Lequin, Clin Chem. 51 (2005) 2415.
[3] B. Liedberg, I. Lundstrom, and E. Stenberg, Sens. Actuat. B I I (1993) 63.
[4] J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao, and R. P. Duyne, Nat. Mat.
7 (2008) 442.
[5] R. Karlsson, J. Mol. Recog. 17 (2004) 151.
[6] C. Caucheteur, Y. Shevchenko, L. Shao, M. Wuilpart, and J. Albert, Opt. Express. 19
(2011) 1656.
[7] S. John, Phys. Rev. Lett. 58 (1987) 2486.
[8] E.Yablonovitch, Phys. Rev. Lett. 58 (1987) 2059.
7
Chapter 2 Si photonic crystal resonator biosensor
2.1 Finite-Difference-Time-Domain analysis of the photonic crystal resonators
The finite-difference-time-domain1) (FDTD) method and the refractive index
approach were used for the simulation of photonic crystal resonator device including the
band diagrams calculation, characteristics of resonant frequency, field distribution and
Q factor. The entire simulation process was carried out using RSoft
FullWAVE™simulation software. The lattice parameters (such as lattice constant a, air
hole radius r) of the PhC structure were varied to fit the resonant peak within the 1.3 m
wavelength.
2.2 Designing of photonic crystal resonator The lattice constant a, and air hole radius r are the key parameters of the photonic
crystal resonator based device. After confirming the lattice constant, then air hole radius
is calculated by r = 0.3a. The input/output waveguide are created by omitting a single
row PhC air holes and width of waveguide should be about . To design the proposed
device, we were used L-Edit v16.0 layout. Figure 2.1 shows the layout of device
structure.
a 3
8
.
Fig. 2.1. Design layout of PhC cavity resonator.
Input
Drop port
Through port
Resonator
Si-waveguideSi-waveguide
Si-waveguide
9
2r
Outdrop
Input
Cavity
a
Outthrough
2.2.1 Cavity type resonator The cavity type resonator was made by increasing the radius of the center air hole
of hexagonal lattice structure. The large resonator area helps to maximize the light and
biomolecule interaction. The Schematic of cavity type PhC resonator is shown in Fig.
2.2.
Fig. 2.2. Schematic of cavity type PhC resonator.9)
2.2.2 Defect type resonator
The Schematic of cavity type2-8) PhC resonator is shown in Fig. 2.3. The defect is
made by omitting the center air hole of structure. However, in the defect type structure
only evanescent light is in the resonator region and it interacts with biomolecules.
10
2r
Outdrop
Input
Defect
a
Outthrough
2r
Outdrop
Input
Cavity
a
Outthrough
Fig. 2.3. Schematic structure of defect type PhC resonator.9)
2.2.3 Special shape resonator (cavity connected with surrounding air
holes)
As the design values of any structures substantially increase a bit after the electron
beam lithography. The special shape structure is just the connection of surrounding
holes and cavity each other. The schematic structure is shown in Fig. 2.4.
Fig. 2.4. Schematic structure of special shape PhC resonator.
11
2r
Outdrop
Input
Adsorption area
a
Outthrough
2.2.4 Double cavity resonators with neighbor hole modulation
Fig. 2.5. Schematic structure of double cavity PhC resonator.
The double cavity resonator with neighbor hole modulation is shown in Fig. 2.5. As
for the large cavity resonator area, much light-biomaterial interaction is possible
because electric field distribution is maximum in the modulated air hole region.
2.3 Resonance characteristics of the device
As discussed in the above for both the structures of the cavity and defect-type PhC
resonators, the sensing depends on the cavity and defect areas and the effective
refractive index change induced by adsorbed biomaterials. The device operation is
explained by the following equation:10)
where λres is the resonance wavelength, neff is the effective refractive index, L is the size
of the resonator, and m (= 1 in the fundamental mode) is an integer. The resonator size
L seems to be roughly equal to the length of the defect region for the defect-type; for the
cavity-type, L may be the sum of the diameter of the center hole and the size of the
remaining defect region.
(2.1) m
Lneffres
12
The shift of resonance wavelength res when ambient refractive index changed and is
expressed by the following equation:10)
where neff is the effective refracitve index of cavity/defect region.
2.4 Details about Q-factor, sensitivity, figure of merit and detection
limit Q-factor:
The quality factor Q of resonation, which is defined by the following equation:
where is the resonance wavelength and FWHM is the full width at half maximum of the
resonance wavelength.
Sensitivity: The sensitivity of PhC resonator biosensor is calculated by following equation:
where res is the shift of resonance peak and n is the change of ambient refractive
index of resonator area.
Figure of merit (FOM): The figure of merit is defined by the following equation:
where S the sensitivity, Q is the quality factor and res is the resonance wavelength.
(2.2)
(2.4)
(2.3)
(2.5)
eff
effresres n
n
FWHMQ res
nS res
res
QSFOM
13
E2 of lightPhC Biomaterials
PhC
E2 of light
Biomaterials
Cavity type
Defect type
PhC
PhC
Detection limit (DL): In order to analyze the sensor performance quantitatively, the detection limit11) of the
PhC resonator sensor is calculated by the sensitivity S, quality factor Q and the
resonance wavelength. The equation is as follows:
2.5 Comparison of Q value between cavity and defect type structure
Fig. 2.6. Schematic of light matter interaction in cavity and defect type.
The front schematic view of cavity and defect type is shown in Fig. 2.6. In the
defect type, the evanescent part of light interacts with biomaterials, whereas cavity light
intensity large inside the cavity region. As for the concentrated electric field, the high Q
and larger resonance wavelength shift is possible by cavity type device.
(2.6) .QS
DL res
14
References [1] J. D. Joannopoulos, S. G. Johnson, J. N. Winn, and R. D. Meade, “Photonic crystal:
Molding the flow of light” Princeton University Press, 2007, ISBN:978-0-691-
12456-8.
[2] X. Fan, I. M. White, S. I. Shopova, H. Zhu, J. D. Suter, and Y. Sun, Anal. Chem.
Act. 620 (2008) 8.
[3] S. Olyaee and A. Naraghi, Phot. Sens. 3 (2013)131.
[4] S. Olyaee, A. Naraghi, and V. Ahmadi, Optik 125 (2014) 596.
[5] S. Olyaee and A. A. Dehghani, Phot. Sens. 2 (2012) 92.
[6] S. Olyaee and A. A. Dehghani, Sens. Lett. 11 (2013)1854.
[7] S. Olyaee and M. Azizi, Phot. Sens. 4 (2014) 220.
[8] S. Olyaee and A. M. Bahabady, Sens. Lett. 13 (2015) 1.
[9] A. K. Sana, K. Honzawa, Y.Amemiya, and S. Yokoyama, Jpn. J. of Appl. Phys. 55
(2016) 04EM11.
[10] M. Fukuyama, S. Yamatogi, H. Ding, M. Nishida, C. Kawamoto, Y. Amemiya, T.
Ikeda, T. Noda, S. Kawamoto, K. Ono, A. Kuroda, and S. Yokoyama, Jpn. J. of
Appl. Phys. 49 (2010) 04DL09.
[11] I. M. White and X. Fan, Opt. Express 16 (2008) 1020.
15
Chapter 3 Fabrication and measurements of photonic crystal resonators
3.1 Fabrication of Si photonic crystal resonator biosensor The fabrication process of the PhC resonator sensor is shown in Fig. 3.1. (1) A 100-
nm-thick oxide hard mask layer was first thermally grown on the SOI wafer surface as
an intermediate layer of pattern transfer. (2) An electron beam (EB)-sensitive positive
photoresist, ZEP520A, was spin-coated on the oxidized SOI wafer surface, and the
photonic crystal hole pattern and waveguide were written using the Elionix ELS-G100
electron beam lithography system, which is of the point beam type. The pattern was
then developed using xylene and isopropyl alcohol (IPA), the second lithography was
carried out to form the PhC holes and cavity. (3) Then the sample with the oxide hard
mask was dry-etched by reactive-ion etching (RIE) using CF4 gas to transfer the pattern
to the hard mask. (4) The patterns were the transferred to the device layer by inductively
coupled plasma (ICP) etching of silicon using Cl2 gas. After completion of the above
process, wet etching was carried out using diluted HF solution to the remove the hard
mask.
16
ZEP-520ASi
SiO2
Si (0.3 m)SiO2(1.1 m)
Si (250 m)
SiO2 hard mask (0.1 m)
(3)
(4)(2)
(1)
Fig. 3.1. Fabrication process of 2D photonic-crystal-based biosensor.1)
A comparison between two electron beam lithography machines (Hitachi HL-700,
variable-shape beam type; Elionix ELS-G100, point beam type) is shown in Figs. 3.2(a)
and 3.2(b), which show that the scattering of hole size in the case of using the ELS-
G100 is much improved. In HL-700, the hole size was deformed owing its variable
Steps of device fabrication
◆Thermal oxidation (100 nm)
◆Electron beam lithography (for waveguide)
◆Resist coating (ZEP520A)
◆Electron beam lithography (for PhC and cavity)
◆Inductively coupled plasma etching of Si (using Cl2 gas)
◆Pattern developed using xylene and IPA
◆Pattern developed using xylene and IPA
◆Reactive-ion etching of SiO2 (using CF4 gas)
17
shaped electron beam. On the other hand, in the case of using the ELS-G100 point beam
system, the hole shape is improved similarly to the design.
Fig. 3.2. Comparison of scattering of fine-tuned PhC hole fabrication between (a) HL-
700 and (b) ELS-G100 electron beam lithography systems.1)
3.1.1 Reason for high Q by proposed device structures In our proposed device structures, we got improved Q for both cavity, special shape
cavity and double cavity with neighbor hole modulation that for point beam lithography
systems (ELS-G100). The point beam lithography much improves the smoothness of
the photonic crystal air holes and cavity region as well.
(b) (a)
Hole diameter (nm)
Freq
uenc
y
Gaussian Approximation
HL700Variable-shape beam
18
Scanning electron micrography (SEM) images of the fabricated cavity (400 nm), defect,
and cavity (450 nm) are shown in Figs. 3.3(a)-3.3(c), respectively.
Fig. 3.3. SEM images of the fabricated devices: (a) cavity type (430 nm)1) (b)
cavity-type (450 nm)1) where the center hole is connected to the
surrounding holes, (c) defect-type1) and (d) double cavity type (457 nm)
where neighbor hole is modulated and connected with center cavity.2)
(a)
1m
Cavity
Double Cavity
1m
(d)
(b)
Wav e
put
WaveguideInput
Drop Port
Cavity
1m
Through Port
1m
Defect
(c)
19
Tunable laser
=1280-1320 nm
Detector
Sample chip
Lens
Optical power meter
PhC pattern
Lensed fiber
Lensed fiber
3.2 Measurement system of Si photonic crystal resonator biosensor 3.2.1 Schematic of measurement systems The schematic of optical measurements systems for the photonic crystal resonator
biosensor is shown in Fig. 3.4. The light from the tunable laser source was coupled into
the waveguide of the device by a lensed fiber, and the output was measured using an
InGaAs photodetector.
Fig. 3.4. Schematic of optical measurement sytems.1)
20
Stage
Fiber
Sample
3 cm
The photograph of the measurement system is shown in Fig. 3.5.
Fig. 3.5. Photograph of the optical measurements system.
21
References [1] A. K. Sana, K. Honzawa, Y.Amemiya, and S. Yokoyama, Jpn. J. of Appl. Phys. 55
(2016) 04EM11.
[2] A. K. Sana, Y. Amemiya, and S. Yokoyama, “High sensitivity double-cavity silicon
photonic crystal resonator for label free biosensor,” Jpn. J. of Appl. Phys. 56 (2017)
(Accepted for publication in 2017).
22
Chapter 4 Detection of device performance using sucrose solution
4.1 Sucrose solution preparation
The way to prepare sucrose solution is shown in Fig. 4.1. Deionized water (DI) of
900 L and 0.1 gm sugar were mixed to make 10% of sucrose solution. After that, 0.1%
sucrose solution was prepared by mixing with 900 L DI water and 10 L of 10%
sucrose solution. Same way 0.2 % was prepared by mixing with 800 L of DI water
with 200 L of 10% sucrose solution.
Fig. 4.1. Steps to prepare sucrose solution.
DI water (900 L) + Sugar (0.1 gm) =10 % solution
DI water (900 L) + 100 L of 10 % of sucrose solution = 0.1 % solution
DI water (800 L) + 200 L of 10 % of sucrose solution = 0.2 % solution
23
4.2 Measurement procedures Firstly, for the sucrose detection, the device was cleaned by H2SO4+H2O2 solution
after that we started measure sucrose solution at different concentration from lower to
higher order. For the each concentration a 20 L solutions were dropped to the sensor
surface and wait for 10 min. After the waiting time we measured the resonant spectra
for each concentration separately. The steps of experimental procedures are shown in
Fig. 4.2.
Fig. 4.2. Measurement steps of PhC resonator using sucrose solution.
After measuring each concentration, device should be cleaned at least four times by
DI water.
◆ SH cleaning by H2SO4+H2O2 (3:1) for 10 min.
◆ Rinse with Deionized (DI) water for 8 min.
◆ Drop sucrose solution by micropipette and wait for 10 min.
◆ Measure the resonance spectra
24
1.28
1.30
1.32
400 450 500
Lattice constant, a =330 nmPhC hole radius, r =0.32a
Res
onan
ce w
avel
engt
h (
m)
Cavity hole diameter (nm)
Simulation
Experimental Cavity
4.3 Details about sucrose concentration and its corresponding refractive index
4.3.1 Measurement results First, for the cavity-type resonator, the relationship between resonation wavelength
and cavity size was simulated. The result is shown in Fig. 4.3 together with the
experimental results. Cavity size has an important role in obtaining a resonance peak
within the 1.3 m range. To prevent water absorption loss, first of all, we consider that
our optical power meter range lies between 1.28 and 1.32 m. To obtain resonance
wavelengths within this range, cavity size should also be considered in the design. The
measured results are close to the simulated results.
Fig. 4.3. Change in resonance wavelength with respect to cavity hole diameter.1)
Figure 4.4 shows the simulated results of the cavity diameter dependence of the
resonance wavelength shift at 0.1% sucrose concentration. The simulation results for the
defect-type resonator (one hole missing) and the measured results for the cavity-type
resonator with an abnormal shape of the cavity (will be discussed later) are also shown.
The resonance wavelength shift becomes large with increasing cavity size. The
25
0
0.2
0.4
1310.5 1310.75 1311
Resonance wavelength (nm)
Out
put (
nW)
Sucrose solution
0% 0.1% 0.2%
Q 105
FWHM0.011 nm
Cavity hole diameter 430 nm
phenomenon clearly shows why a large cavity offers a larger wavelength shift than
other structures. The large cavity is filled up with the highest amount of target
molecules, as a result the confined photons interact markedly with the molecules; finally,
the wavelength shift becomes large.
Fig. 4.4. Resonance wavelength shift () sucrose concentration with respect to
cavity hole diameter.1)
The measured results for the sucrose solution are shown in Figs. 4.5 and 4.6,
respectively, for the normal cavity hole (diameter of 430 nm) and abnormal cavity hole
(accidentally a too large hole was formed and it is connected to the surrounding holes).
We measured the output of the through port because the output of the drop port was
weak for some reason (for example, caused by the defects in the waveguide connecting
to the drop port). We observed clear dips due to the resonation. It is clearly observed
that the resonance wavelength has shifted with sucrose concentration change.
26
0.001
0.01
0.1
1
10
300 350 400 450
Cavity type(Simulation)
Defect type (Simulation)
Cavity type (Meas.)center hole > 450 nm
Cavity hole diameter (nm)
su
cros
e co
ncen
tratio
n (n
m/%
)
Fig. 4.5. Measured results of fabricated device with cavity hole diameter 430 nm.1)
27
0
0.2
0.4
1296.5 1297.0 1297.5
Sucrose solution
0% 0.1%0.2%
Q 105
FWHM0.01 nm
Resonance wavelength (nm)
Out
put (
nW)
Cavity hole diameter 450 nm
Fig. 4.6. Measured results of fabricated device with cavity hole diameter 450 nm.1)
In the double nanocavity resonator type device the neighbor hole radius are varied
and we got highest Q-factor at certain radius. The experimental result of double
nanocavity type resonator at various sucrose concentrations is shown in Fig. 4.7. The
observed resonance peaks have very small full width at half maximum (FWMH) of 6.5
to 6.9 pm and the Q-factor about (1.93 ~ 2.02) 105 is derived from the various FWHM.
28
0.00
0.10
0.20
0.30
1316.2 1316.4 1316.6 1316.8Wavelength (nm)
Out
put (
arb.
u.) Q (1.93~2.02) x 105
Sucrose solution
0% 0.1% 0.2%
FWHM6.9 pm
FWHM6.5 pm
FWHM6.7 pm
(a)
Average base line
Fig. 4.7. Measured results double nanocavity resonator.2)
29
Low
HighE2
(d)Low
HighE2
(c)
The simulation result of Q-factor versus neighbor hole radius is shown in Fig.
4.8(a) together with the experimental data. The results show that the Q value becomes
high at certain neighbor holes radius.
Fig. 4.8. (a) Neighbor hole radius versus Q value, (b) neighbor hole radius versus
resonaonce wavelength, (c) electric field distribution at neighbor hole 65
nm and (d) electric field distribution at neighbor hole 70 nm.2)
In Fig. 4.8(b) the simulated and experimentally observed resonance wavelengths
are plotted. The deviation between simulation and experiment is larger at the larger
neighbor hole sizes. This trend may be explained by the proximity effect during the
electron beam lithography process
Figure 4.8(c) and Fig. 4.8(d) it is recognized that the light intensity is strong at the
neighbor holes, then its size strongly affects the resonance characteristics, but the
detailed physical mechanism of the enhancement of Q-factor is not clear at this stage.
Neighbor hole radius (nm)
0
1
2
3
50 55 60 65 70 75
Simulation
Experimental
(a)
Q-fa
ctor
(105 )
Neighbor hole radius (nm)R
eson
ance
Wav
elen
gth
(nm
)50 55 60 65 70 75
Simulation
Experimental
1280
1300
1320
1340
(b)
30
0
0.2
0.4
0 0.1 0.2
Defect type(Simulation)
Defect type(Experimental)
Simulation
Cavity 450 nm
Experimental
SimulationExperimental
Cavity 430 nm
Sucrose concentration (%)
Wav
elen
gth
shift
(nm
)
1.33299 1.33313 1.33327Refractive index
4.4 Linearity check with resonant wavelength shift by the sucrose concentration
As we measured our device performances using sucrose solution and we took
attention that resonance wavelength shift should be linear with sucrose concentration. In
Fig. 4.9, for the abnormal cavity shape, the wavelength shift does not linearly change
with increasing sucrose concentration. The possible reason is that air bubbles are
adsorbed on the sidewall of the cavity. The results of two different cavity hole size
structures show that, the cavity with a large hole induces a greater resonance
wavelength shift than that with a small one. In our measured results, the abnormal
cavity shape (designed cavity size is 450 nm) has about 1.7 times larger wavelength
shift at 0.1% sucrose concentration than the small cavity (430 nm). We confirmed a
large 2.3 times wavelength shift of the large-cavity- type structure compared with that
of the defect-type at 0.1% sucrose concentration.
Fig. 4.9. Sucrose concentration vs resonance wavelength shift.1)
31
0
0.30
0.60
0 0.1 0.2Sucrose concentration (%)
1.33299 1.33313 1.33327Refractive index
Simulation (Cd= 400 nm)
Simulation (Cd= 450 nm)
Measured (Cd > 450 nm)
Wav
elen
gth
shift
,
(nm
)
The sucrose concentration versus resonance wavelength shift for double nanocavity
resonator is shown in Fig. 4.10.
Fig. 4.10. Sucrose concentration vs resonance wavelength shift.2)
32
References [1] A. K. Sana, K. Honzawa, Y.Amemiya, and S. Yokoyama, Jpn. J. of Appl. Phys. 55
(2016) 04EM11.
[2] A. K. Sana, Y. Amemiya, and S. Yokoyama, “High sensitivity double-cavity silicon
photonic crystal resonator for label free biosensor,” Jpn. J. of Appl. Phys. 56 (2017)
(Accepted for publication in 2017).
33
Chapter 5 Detection of antigen-antibody reaction
5.1 Details of Si-tag, protein-G, antibody (IgG2a) and prostate specific antigen
Rapid and early diagnosis is demanded for the aging society1) to providing better
quality of life as well as to minimize the healthcare expenses.2) Conventional biosensors
based on enzyme-linked immunosorbent assays (ELISA)3, 4 ) requires a number of steps,
including the use of a labeling target molecules to get a detectable signal. A lot of study
has been devoted to the developed the fast, accurate, and label-free biosensors to
enhance medical diagnostics, biomolecules, and organic chemical detection.5-9) Photonic
crystal10-12) based micro/nano cavities getting lots of research interest for medical
diagnosis purpose due to their strong light confinement nature, high quality factors (Q)
and small modal volumes. By introducing a defect in the photonic crystal periodic
structure, the light of certain resonance frequency are possible to confine into the defect
or cavity region. When the local refractive index of the sensor surface is changed then
resonant wavelength shifts. These shifts are the main key parameter to measure the
device sensitivity
The uniqueness of this work is the silicon binding protein that binds with SiO2 and
Si surface and immobilizes the biomolecules on the PhC cavity surface. The things that
we used are the protein G (Pro-G) which makes strong bonds with many kinds of
mammalian antibodies. The antigen-antibody reaction procedure is shown in Fig. 5.1. In
Fig. 5.1(a), the receptors are randomly oriented. In this method antigen-antibody
immobilization on sensor surface is difficult. In Fig 5.1(b), Si-tag were used to
immobilize the antigen-antibody on the sensors surface and role of Pro-G is shown in
Fig. 5.1(c). The PhC nano-cavity resonator is immersed in the solution containing the
Si-tagged Pro-G, which immobilizes the bioreceptors on the Si/SiO2 surface, and then
immersed in the antibody (mouse antibody subtype IgG2a) solution. After that, the
resonator is immersed by the target antigen of PSA. The resonance spectra have been
measured at each step.
34
SiNSi
PhC-hole Receptor
SiNSiSi-tagPro-G
Antibody (IgG2a)PSA
(a) (b)SiN
Si-tag
Si(c)
Fig. 5.1. Role of Si-tag (a) without Si-tag (b) with Si-tag and (c) role of Protein G.13, 14)
5.2 Measurement procedures
The measuring steps of PSA are explained in Fig. 5.2. First, the Si-tagged protein
G was poured on the surface sensor, and then the resonance spectrum was measured.
Secondly, the anti-PSA or antibody (named as IgG2a) was added to the solution, the
resonance wavelength was shifted. Next, PSA antigens were added to the solution, and
measured the spectral shift of resonance peak. Buffer solution (Tris-HCl) was used to
remove the unreacted biomolecules from the sensor surface. The resonance spectra for
various concentrations were measured from lower to higher concentration.
35
Fig. 5.2. Measurement steps of PhC resonator using antigen-antibody reaction.
◆Drop 20 μL of Si-tag + Pro-G and wait for 10 min
◆Clean by Tris-HCl ◆Drop 20 μL of IgG2a (Antibody) and wait for 10 min
Steps (2) should be repeated from lower concentration to higher concentration.
◆Measurements
◆Clean by Tris-HCl
◆Measurements
◆SH cleaning by H2SO4+H2O2 (3:1) for 10 min
◆Sensor surface clean by Buffer solution (Tris-HCl) and wait for 10 min
(1)
◆Rinse with Deionized (DI) water for 8 min
◆Clean by Tris-HCl ◆Drop 20 μL of PSA(antigen) and wait for 10 min
(2)
36
0
0.1
0.2
0.3
0.4
0.5
1311.8 1312 1312.2 1312.40
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
0
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
0
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
0
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.40
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4-0.02
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
-0.02
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
Si-Tag+Pro G
0.1 ng/mL
0.3 ng/mL
0.5 ng/mL 2 ng/mL
3 ng/mL
Wavelength (nm)
Out
put (
nW)
(a)
0
0.1
0.2
0.3
0.4
0.5
1311.8 1311.9 1312 1312.1 1312.2 1312.3 1312.4
0.01 ng/mL
1 ng/mL
.0
Fig. 5.3. Experimental results of different PSA conentration.
5.3 Resonance characteristics versus PSA concentrations Finally, we measured prostate specific antigen (PSA) biomarker. In many cases,
concentration of PSA of 4 ng/mL in the blood is considered the prostate cancer positive.
The sensitivity of the biosensor is required to detect less than 1 ng/mL. The
experimental procedure is explained in Fig. 5.2. First, the Si-tagged protein G was
poured on the surface sensor, and then the resonance spectrum was measured. Secondly,
the anti-PSA or antibody (named as IgG2a) was added to the solution, the resonance
wavelength was shifted. Next, PSA antigens were added to the solution, and measured
the spectral shift of resonance peak. Figure 5.4 shows the resonance spectra of Si-
tag+proG and several of PSA concentrations and it also shows that at each
concentration resonance spectra has shifted and at higher concentration the shift is going
to be saturated.
37
0
0.10
0.20
0.30
0.40
0.50
10 -12 10 -11 10 -10 10 -9 10 -8
Wav
elen
gth
shift
(nm
)
PSA concentration C (g/mL)
CKCK
1max
46.0max 10103K
(b)
CKCK
1max
5.4 Measurements results fitting with Langmuir’s curve
In Fig. 5.3 the resonance wavelength shift (Δλ) is plotted as a function of the concentration of the PSA solution. The result fits to the following Langmuir equation,15)
where Δmax is the maximum resonance wavelength shift, C is the concentration of
PSA biomarkers and K indicates the equilibrium constant of adsorption-desorption
reaction. The best fitting is shown in Fig. 5.4 by adjusting the K values.
Fig. 5.4. Langmuir’s fitting with experimental results.
5.5 How could linearity improvement is possible in saturated region at higher concentration of biomaterials
As in Fig. 4.9, the resonance wavelength shift of various device structures are
shown. Both cavity and special shape cavity has larger wavelength shift and these two
structures will saturate quickly at higher concentration. However in the defect type
structures, the shift is smaller than other structures. As for the small shift, the linearity at
higher concentration may be possible by using defect type device.
(5.1) ,
38
References [1] G.-J. Zhang, Z. H. Luo, M. J. Huang, J. J. Ang, T. G. Kang, and H. Ji, Bios. Bioele.
28 (2011) 459.
[2] T. Taniguchi, A. Hirowatari, T. Ikeda, M. Fukuyama, Y. Amemiya, A. Kuroda, and
S. Yokoyama, Opt. Comm. 365 (2016) 16.
[3] C. Liu, F. Wang, T. Sun, Q. Liu, H. Mu, and K. P. Chu, J. Nanophot. 9 (2015).
093050(1-10).
[4] R. S.Yalow and S. A. Berson, Clin. Invest. 39 (1960) 1157.
[5] E. K. Akowuah, T. Gorman, H. Ademgil, S. Haxha, G. K. Robinson, and V. Oliver,
IEEE J. of Quan. Elect. 48 (2012) 1403.
[6] J. A. Kim, T. Hwang, S. R. Dugasani, R. Amin, A. Kulkarni, S. H. Park, and T. Kim,
Sens. Actu. B 187 (2013) 426.
[7] J. Homola, Anal. Bioanal. Chem. 377 (2003) 528.
[8] R. Otupiri, E. K. Akowuah, S. Haxha, H. Ademgil, F. Abdelmalek, and A. Aggoun,
IEEE Phot. J. 6 (2014) 6801711.
[9] G. J. Ortega-Medoza, A. P. Vivanco, C. T. Quitl, P. Z. Morán, V. Hernández, and F.
Chávez, Sens. 14 (2014) 18701.
[10] E. Chow, A. Grot, L. W. Mirkarimi, M. Sigalas, and G. Girolami, Opt. Lett. 29
(2004) 1093.
[11] M. Lee and P. M. Fauchet, Opt. Express 15 (2007) 4530.
[12] S. Hachuda, K. Watanabe, D. Takahashi, and T. Baba, Opt. Express. 24 (2016)
12886.
[13] T. Ikeda, Y. Hata, K. Ninomiya, Y. Ikura, K. Takeguchi, S. Aoyagi, R. Hirota, and
A. Kuroda, Anal. Biochem. 385 (2009) 132.
[14] K. Taniguchi, K. Nomura, Y. Hata, T. Nishimura, Y. Asami, and A. Kuroda,
Biotechnol. Bioeng. 96 (2007) 1023.
[15] I. Langmuir, J. Am. Chem. Soc. 38 (1916) 2221.
39
Chapter 6 Temperature independent device for the biomaterial sensing
6.1 Temperature effects on resonance wavelength shift
The refractive index (RI) of silicon greatly changes with temperature change.1,2)
The effective refractive index of the sensor surface (Si ring resonator, PhC cavity
resonator or any other RI based sensor) changes due to the adsorption target material.
Then it becomes difficult to measure target material induced resonance wavelength shift
precisely because of the thermal fluctuation changes the silicon refractive index as well
as takes part on the resonance wavelength shift. Controlling the thermal effects is
important to detect the biomaterials perfectly as the effective refractive changes with
temperature change. The change in biomolecules concentration induces the change in
local refractive index as a result resonant wavelength shift occurs. However the silicon
itself has large thermo-optic coefficient3) (dn/dT=1.810-4/C, where n is refractive
index and T is temperature). Furthermore deformation of the device mechanical shape
due to temperature change induces the resonance wavelength. Due to large thermo-optic
coefficient of Si, active thermal controller is needed to detect the biomolecules precisely.
However, for the low cost applications, active thermal control is often not feasible.
Several thermal effect compensating devices so that they can be used without any
thermal controller, have previously been shown4-19) by designing the specific device
structure that compensate thermal effects. As the temperature changes, the device
mechanical deformations happen together with the refractive index change of Si.
40
Si-sub
SiSiO2
LSi
L
Si-subTdSiO2
dSi
R
SiO2
,)(1
)1)(1()1(3
)1(622
221
pqppqph
pTR
Fig. 6.1. Curvature shape of thin Si layer due to mechanical deformation by
temperature change.
6.2 Numerical analysis and simulation
Resonance wavelength shift of Si resonators by thermal change is given by the
following equation,
where res, neff and L are the resonance wavelength, effective refractive index and size
of the resonator respectively. res, neff and L are the change in these values due to
the temperature change. The right hand first term indicates the effective refractive index
(Si) effect and latter term indicates the thermal deformation effect. This equation is
easily derived from the equation of res= neffL/m, where m denotes the integer. As the
temperature increases, Si and SiO2 thermally expand with different thermal expansion
coefficients and the SOI wafer has a curvature shape as shown in Fig. 6.1. The radius of
the curvature of the SOI substrate is given by the following equation,22)
where h, 1 and 2 are the total thickness and thermal expansion coefficient of Si and
SiO2, p=dSiO2/dSi (where dSiO2 and dSi are the thickness of SiO2 and Si, respectively),
q = YSiO2/YSi (where YSiO2 and YSi are Young’s modulus of SiO2 and Si) and ΔT is the
temperature change. The change in resonator size is expressed as follows
(6.1)
(6.2)
, resreseff
effres L
Lnn
41
,2RdLL
where d is the thickness of SOI substrate. Here, d=dSi+dSiO2 and dSi >> dSiO2 is assumed.
In the differential silicon ring resonator, one ring is considered as detection ring, on
which a biomaterial is adsorbed and other is considered as reference ring, which is not
exposed to the biomaterial. The phase of one of the outputs is changed by π and merged
again. When a biomaterial is not adsorbed on the detection ring, the differential output
is zero because these two resonance curves completely overlap. When a biomaterial is
adsorbed on the detection ring, its resonance curve shifts and an output signal appears.
Here, we discuss the output spectrum of the differential ring resonator sensor. First, for
the single-ring resonator the transmission spectrum TS is given by the following
equations24, 25)
where sin K is the coupling coefficient, is the attenuation constant of the waveguide,
L is the peripheral length of the resonator, and neff is the effective refractive index of Si
waveguide. The transmission spectrum of the differential Si ring resonator sensor TDS is
given by20)
where is the phase difference between the detection ring and reference ring and usually .
The temperature dependence is calculated by using following equations
(6.3)
,)(cos21
)](sin)(cos1[sin2
XXiXKTs
(6.4)
),2
exp(cos2 LKX (6.5)
,2
)(
Lneff (6.6)
,2)cos22(2
sTsTiesTDST (6.7)
42
where l is the length of the phase shifter, T ( ̊C ) is the temperature, n23 is the effective
refractive index at 23 C , and is thermal coefficient of core and various clad layer
and at the above temperature . n23 and are obtained by using the following equation.
where n1, n2, n3 are refractive index of core, top clad and bottom clad respectively. the
thermal coefficient of core, top clad and bottom clad are dn1/dT, dn2/dT and dn3/dT
respectively. 1 , 2 and 3 are the electric field confinement factor of core, top clad and
bottom clad respectively, which were obtained from the simulation using Rsoft
photonic CAD suite (Synopsys Inc.). The equation (6.9) is obtained by optical simulator.
Using the values of the bulk materials are obtained from Refs. 26 for oil , and 27 for
silicon, SiO2, resist and air. The actual numerical values of coefficient = 1.16 10-4/
C and for SiO2 and = 1.14 10-6/ C for water, = 1.09 10-4/ C for air, = 1.15
10-4/ C for oil, and = 1.15 10-4/ C for resist respectively.
In the differential mode, wide bandwidth input light can be used and the output is the
integral of the light intensity within the input light bandwidth or the maximum peak
height.
Figure 6.2 shows the maximum differential output versus neff /neff or ΔL/L which is
difference in the effective refractive index or circumferential length of Si waveguide
between the detection and reference resonators divided by the original ones respectively.
It is found that there is a linear region in the output versus neff /neff or ΔL/L curve. The
neff or ΔL induces res which is difference in the resonance wavelength between the
two rings. In the linear region even though the resonance wavelengths of both
resonators are slightly different, the sensitivity which defined as output/neff or
output/res, becomes constant when the operation point is in the linear region.20)
(6.9)
(6.8)
(6.10)
)23(23
Tnneff
,2
lneff
,33
322
211
1bottomcladtopcladcore
effeff dT
dnndTdnn
dTdnn
dTdn
n
43
Figures 6.3 and 6.4 shows the calculated temperature change versus maximum output
change for various cladding layers (e.g. air, water, resist, oil, and SiO2).
To simulate the temperature dependence of photonic crystal resonator sensor we used
equation (6.1)-(6.3) and equation (6.10). The length L for photonic crystal resonator is
considered the maximum electric field distributed area. The field is distributed in the Si,
resonator inside as well as in the SiO2 layer. The confinement factor 1, 2 and 3 are
for the inside of the resonator area, Si and SiO2 respectively. The simulated electric field
distribution using the Rsoft photonic CAD suite is 58%, 22% and 20% of inside of the
resonator region, Si and SiO2 respectively.
44
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.0 0.0 0.0 0.0Max
imum
out
put
(rel
. uni
t)
Refractive index change0 2 4 6 8
0
0.2
0.4
0.6
0.8
1.0
Diff
eren
tial o
utpu
t (re
l. un
it)
Operation point
Scattering of res
(a)neff/neff or L/L (10-5)
0.3962
0.3967
0.3972
0.3977
0.3982
19 21 23 25 27Temperature T (C)
Max
imum
diff
eren
tial o
utpu
t(re
l. un
it)
(b)
Airwater
Fig. 6.2. Simulation results of refractive index versus maximum output change.
Fig. 6.3. Simulation results of temperature change versus maximum differential output variation at various cladding layer such as air and water.
45
0.3962
0.3967
0.3972
0.3977
0.3982
19 21 23 25 27
Max
imum
diff
eren
tial o
utpu
t(re
l. un
it)
(c)Temperature T (C)
ResistOilSiO2
Fig. 6.4. Simulation results of temperature change versus maximum differential output variation at various cladding layer such as resist, oil and SiO2.
6.3 Schematic of device structures of differential Si ring resonators
In this work, the resonance wavelength shifts of Si ring and photonic crystal
resonators on silicon-on-insulator (SOI) substrate were investigated. In order to solve
the temperature instability we proposed a differential operation of two resonators,20)
where a phase shifter is installed in one of the two outputs before merging. In this case
the integral intensity of the differential output becomes temperature independent
because the resonance wavelengths of both two resonators shift in parallel. The
schematic structure of the differential Si ring resonator and photonic crystal (PhC) based
cavity type resonators21) are shown in Figs. 6.5(a) and 6.5(b) respectively.
46
Output
Input
Cavity
Period : p
Bio-materials
(b)
Input Output phase shifter
Reference ring
Sensing ringBiomaterials
(a)
Fig. 6.5. (a) Schematic of differential Si ring resonator biosensor and (b) photonic crystal cavity resonator.
47
6.4 Measured results
6.4.1 Mechanism of thermal change of res of Si ring and PhC resonators
The temperature dependence of the differential Si ring and PhC cavity resonators
are shown in Figs. 6.6 and 6.7 respectively. In Fig. 6.8(a) the measured results for the
differential Si ring resonators at different temperature are shown together with the
simulation results obtained by the method described in Sect. 6.2. It is found that the
measured data well fits to the simulation result and it is clear that Si refractive index
change effect is dominating. The mechanical deformation effect is negligible (res/T~
-10-4 nm/C). The result for the PhC resonator is also shown in Fig. 6.8(b), and the same
conclusion as for the ring resonator is derived. For the both type of devices, we
simulated by using equation (6.10) and considered core, top clad and bottom clad each
layer’s effect individually. The calculated field distribution in the core (Si), top clad (air)
and bottom (SiO2) clad are 80%, 8% and 12% respectively.
6.4.2 Thermal stability of differential Si ring resonator sensor
In Fig. 6.9 the maximum of differential output change for T =4 ̊C from 25 C to 29
C for the Si ring resonators with different top cladding layers are plotted. In the
differential detection, only the maximum differential output is counted, which
corresponds to the difference between the resonance wavelength shifts of two rings. And
the spectral shift of the differential output is not counted. Theoretically, the maximum
differential output change for T=4 ̊C is calculated at three different top cladding layer
conditions, the simulated change for resist, air and oil are 0.012%, 0.02% and 0.01%
respectively indicated by . The measured results is also shown in Fig. 10, where the
changes in maximum differential outputs for the differential Si ring resonators with
different top cladding layers for T =4 ̊C are plotted. Measured scatter error for multiple
measurements in air clad is represented by (reproducibility at the same condition).
And measured data deviation with the temperature changes (25 C to 29 C) is
expressed by . The theoretical calculation suggests that temperature independence of
the differential operation is right.
48
0
4
8
12
1297.5 1297.9 1298.3 1298.7
Diff
eren
tialo
utpu
t (nW
)
Wavelength (nm)
25 C 27 C29 C
Air-medium
Si ring resonator
In our research we targeted to detect prostate specific antigen (PSA) of 1 ng/mL as
practical sensitivity. We used various concentrations of sucrose solutions and the
conversion of equivalent refractive index was done by the help of Ref. 28 and 29. The
sucrose solution at 10-2 % concertation corresponds to 1 ng/mL of PSA. The change in
effective refractive index for 1 ng/mL PSA is equivalent to Δneff 1.710-5 (Ref. 20)
and Δneff/neff is the 6.510-6, which corresponds to the differential output of 0.186 from
Fig. 6.2. This is on the linear region of this graph. As the differential output is in the
linear region, output/Δλres could be constant even though the temperature increases.
Fig. 6.6. Measurement results of differential Si ring resonator at different temperatures. Top cladding layer is air.
49
0
0.02
0.04
0.06
0.08
1310.5 1310.6 1310.7
Air-medium
Wavelength (nm)
Out
put (
nW)
PhC resonator
25 C26 C
27 C
Fig. 6.7. Resonation spectra of PhC cavity resonator at various temperatures.
50
-0.05
0.1
0.25
0.4
0.55
0.7
0 1 2 3 4
Meas.
Ref. index effect
Ref. index+ Mech. effect
Mech. effect0 0
Si-ring resonator
Temperature change T (C)
Res
onan
ce w
avel
engt
h sh
ift
(nm
)
MeasuredSimulation
(a)
Simulation
-0.01
0.01
0.03
0.05
0.07
0 1 2
0 0
Temperature change T (C)
Res
onan
ce w
avel
engt
h sh
ift
(nm
)
PhC resonatorMeasuredSimulation
Meas.
Ref. index effect
Ref. index+ Mech. effect
Mech. effect
(b)
Simulation
Fig 6.8. Effect of temperature on resonance wavelength shift for (a) Si ring resonator and (b) PhC cavity resonator (where silicon effective refractive index change (Ref. index effect) and mechanical effects (Mech. effect) are abbreviated).
51
0
60
80
100Resist Air Oil
Max
imum
diff
eren
tial o
utpu
t (%
)
120Differential Si ring resonator
(0.012%) (0.01%)(0.02%)
Fig 6.9. Comparison of integrated light output variation in different medium at temperature range between 25 C and 29 C. Measured scatter error for multiple measurements in air clad at 25 C is expressed by . Multiple measured scatter for resist and oil were not carried out. But similar scatter is expected for resist and oil because the scatter occurred for the measurement system error. And measured data deviation with the temperature changes (from 25 C to 29 C) is expressed by . Theoretical calculation for thermal change (for ΔT= 4 C) is represent by and the value were indicates in the parentheses.
52
References [1] G. E. Jellison Jr. and H. H. Burke, J. Appl. Phys. 60 (1986) 841.
[2] J. A. McCaulley, V. M. Donnelly, M. Vernon, and I. Taha, Phys. Rev. B 49 (1994)
7408.
[3] X. Tu, J. Song, T-Y. Liow, M. K. Park, J. Q. Yiying, J. S. Kee, M. Yu, and G-Q. Lo,
Opt. Express 20 (2012) 2640.
[4] J. Teng, P. Dumon, W. Bogaerts, H. Zhang, X. Jian, X. Han, M. Zhao, G. Morthier,
and R. Baets, Opt. Express 17 (2009) 14627.
[5] H. Tanobe, Y. Kondo, Y. Kadota, K. Okamoto, and Y. Yoshikuni, IEEE Photonics
Tech. Lett. 10 (1998) 235.
[6] Y. Kokubun, S. Yoneda, and S. Matsuura, Elect. Lett. 34 (1998) 367.
[7] K. Maru and Y. Abe, Opt. Express 15 (2007) 18351.
[8] T. Claes, J.G. Molera, K. De Vos, E. Schacht, R. Baets, and P. Bienstman, IEEE
Photonics J. 1 (2009) 197.
[9] V. R. Almeida, Q. Xu, C. A. Barrios, and M. Lipson, Opt.Lett. 29 (2004) 1209.
[10] Q. Xu, V. R. Almeida, R. R. Panepucci, and M. Lipson, Opt. Lett. 29 (2004) 1626.
[11] J.-M. Lee, D.-J. Kim, G.-H Kim, O.-K. Kwon, K.-J.Kim, and G. Kim, Opt. Express
16 (2008) 1645.
[12] X. Wang, S. Xiao, W. Zheng, F. Wang, Y. Li, Y. Hao, X. Jiang, M. Wang, and J.
Yang, Opt. Comm. 282 (2009) 2841.
[ 13] L. Zhou, K. Kashiwagi, K. Okamoto, R. Scott, N. Fontaine, D. Ding, V. Akella,
and S. Yoo, Appl. Phys. A: Materials Science & Processing 95 (2009) 1101.
[14] L. Zhou, K. Okamoto, and S. J. B. Yoo, IEEE Photonics Tech. Lett. 21 (2009)
1175.
[15] H. Huang, S.T. Ho, D. Huang, Y. Tu, H. Hu, J. Wang, and W. Liu, J. of Modern
Opt. 57 (2010) 545.
[16] K. B. Gylfason, C. F. Carlborg, A. Kazmierczak, F. Dortu, H. Sohlström, L. Vivien,
C. A. Barrios, W.van der Wijngaart, and G. Stemme, Opt. Express 18 (2010) 3226.
[17] Y. Atsumi, K. Inoue, N. Nishiyama, and S. Arai, Jpn. J. of Appl. Phys. 49 (2010)
50206.
[18] A. H. Harvey, J. S. Gallagher, and J. M. H. L. Sengers, J. of Phys. and Chem. Ref.
53
Data 27 (1998) 761.
[19] L. Eldada, Opt. Engg. 40 (2001) 1165.
[20] T. Taniguchi, Shu. Yokoyama, Y. Amemiya, T. Ikeda, A. Kuroda, and S.
Yokoyama, Jpn. J. Appl. Phys. 55 (2016) 04EM04.
[21] A. K. Sana, K. Honzawa, Y. Amemiya, and S. Yokoyama, Jpn. J. Appl. Phys. 55
(2016) 04EM11.
[22] S. D. Brotherton and T. G Read, Solid State Electron. 13 (1973) 1367.
[23] W. Kern and D. A. Puotinen, RCA Rev. 31 (1970) 187.
[24] Y. Kokubun, Oyo Buturi 72 (2003) 1364 [in Japanese].
[25] Y. Kokubun and K. Kogaku (Optical Wave Engineering) (Kyoritsu, Tokyo,1999) p.
238, 297 [in Japanese].
[26] https://www-s.nist.gov/srmors/tablePDFV.cfm?tableid=181.
[27] http://www.dowcorning.com.cn/zh_CN/content/publishedlit/75-1007-01_single.pdf
[28] R. Budwing, Exp. Fluids 17 (1994) 350.
[29] D. P. Subedi, D. R. Adhikari, U. M. Joshi, H. N. Poudel, and B. Niraula, Kathma.
Univ. J. of Sci., Engg. and Tech. 1 (2006) 1.
[30] A. K. Sana, M. Jun, Shu. Yokoyama, Y. Amemiya, and S. Yokoyama, “Temperature
dependence of resonance characteristics of silicon resonators and thermal-stability
improvement by differential operation,” Jpn. J. of Appl. Phys. 56 (2017) (Accepted
for publication in 2017).
54
Chapter 7 Conclusion
7.1 Conclusion
(1) Silicon photonic crystal resonator biosensors have been developed. We studied
different type of photonic crystal based biosensors such cavity type, defect type and
special type (cavity and surrounding holes are connected each other).
(2) We have demonstrated the excellent potential of light confinement of a 2D photonic
crystal cavity. In particular, the cavity size greatly enhances the large wavelength
shift. We measured highest Q value of >105, which is the highest value ever
reported for any other photonic-crystal-based biosensors. We obtained a great
improve of the sensitivity of 1570 nm=RIU. The proposed cavity-based sensor
structure confirms the resonance wavelength shift with effective refractive index
changes by target molecules. Also, there is a linear dependence between refractive
index changes and wavelength shift.
(3) We confirm that the big cavity-type device has (more than three times) a greater
wavelength shift as than the defect-type device.
(4) We succeeded in sensing the antibody–antigen reaction using photonic crystal
double nanocavity resonator sensor employing PSA marker as an example target. By optimizing the immobilization of the target biomarker,
(5) We detected the PSA concentration as low as 0.01 ng/mL. Thus, our PhC based double nanocavity resonator may be promising candidate as practical biomolecules sensor in the medical diagnosis.
(6) Temperature effects on resonance wavelength shift are studied using both Si ring
and photonic crystal cavity resonator type device.
(7) We reported that the temperature effects on resonance wavelength shift are
dominated by the refractive index change of Si and it is demonstrated that these
temperature effect is sufficiently suppressed by the differential operation. This type
55
of differential operation could also be applied for optical switching, modulation and
so on where thermal effects should be avoided.
(8) Theoretically, we showed that the maximum differential output change for T=4 ̊C
is calculated at three different top cladding layer conditions, theoretically change for
resist, air and oil are 0.012%, 0.02% and 0.01% respectively indicated
7.2 Comparison among other works
In the Table 7.1, we compared our device performance in accordance to sensitivity,
Q-factor and figure of merit with other researcher work. We obtained a greatly improve
the sensitivity of 1570 nm/RIU.
Table 7.1. Performance comparison of our single cavity PhC resonator device with other researcher work.
Sensitivity,
S=n (nm/RIU)
Q=res/FWHM Figure of merit, FOM=
S Q/res
Resonator type
This work 1570 105 1.2 105 Cavity Ref. 1 451 7000 >2000 Defect Ref. 2 60 7000 300 Defect Ref. 3 900 700 400 Defect
56
In the Table 7.2, the device performances are compared with researcher work in
accordance to sensitivity, Q-factor and detection limit. We reported second smallest
detection limit but a very much improved Q-factor.
Sensitivity, n(nm/RIU)
Quality factor, Q=res / FWHM
Detection limit, DL=res/QS (RIU)
Sensor type
This work 1571 2 105 (4.15 ~ 4.34) 10-6 Double-cavity resonator
Ref. 4 500 17890 1 10-4 Defect resonator
Ref. 5 132 2966 3.8 10-6 Cavity resonator
Ref. 6 1500 50000 7.8 10-6 Slot PhC waveguide
Ref. 7 370 7500 2.2 10-4 Cavity resonator
Ref. 8 235 25000 1.1 10-4 Cavity resonator
Table 7.2. Performance comparison of double nanocavity PhC resonator with other researcher work.
57
7.3 Impact of this research work
The work has impact on industry and technology where high sensitivity of detection
is necessary. In the field of biosensor, photonic crystal based biosensor may play
important role where several of kind biomaterial detection is necessary to knowing the
public health.
The designs and demonstration of PhC cavities (cavity/defect and special shape
cavity type) that have been described in this thesis have much more information for
future development of high Q cavities. We studied on the cavity/defect type resonator
and the size of the cavities was also studied and their effects on resonance wavelength
shift. We hope that the various kind device structures certainly help to develop the high
sensitivity device.
By changing the neighbor hole of the cavity or cavities there is a possibility to get
better sensitivity, and high Q-factor as well.
We describe the mechanism and solution for dominating temperature effects on
refractive index based Si optical resonator sensor such as ring resonator and photonic
crystal resonator sensors. The temperature change affects the silicon refractive index
and affects resonator mechanical shape also. As a result, it is reported that the refractive
index change is dominating whereas the mechanical deformation effect is negligible.
We also demonstrated that the differential operation is effective to suppress the
temperature effect for Si ring resonator sensor. This type of differential operation could
also be applied for optical switching, modulation and so on where thermal effects
should be avoided.
The impact of this work on the engineering is the high Q-factor, high sensitivity and
minimum detection limit. On the other hand, the impact on academic are the double
cavity and neighbor hole radius change may have great impact on high sensitivity with
minimum detection limit. We theoretically and experimentally reported that double
cavity connected with radius modulated neighbor holes has great impact on Q value
improvement and sensitivity as well.
58
References [1] B. Wang, M. A. Dundar, R. Ntzel, F. Karouta, S. He, and R. W. Heijden, Appl.
Phys. Lett. 97 (2010) 151105.
[2] L. J. Sherry, S. Chang, G. C. Schatz, and R. P. V. Duyne, Nano Lett. 5 (2005) 2034.
[3] R. Budwing, Exp. in Fluids. 17 (1994) 350.
[4] T. H. P. Chang, J. Vac. Scie. & Tech., 12 (1975) 1271.
[5] J. Zhou, H. Tian, D. Yang, Q. Liu, and Y. Ji, Opt. Comm. 330 (2014) 175.
[6] A. D. Falco, L. O'Faolain, and T. F. Krauss, Appl. Phys. Lett. 94 (2009) 063503.
[7] S. H. Mirsadeghi, E. Schelew, and J. F. Young, Appl. Phys. Lett. 102 (2013)
131115.
[8] C. Caër, S. F. Serna-Otálvaro, W. Zhang, X. L. Roux, and E. Cassan, Opt. Lett. 39
(2014) 5792.
59
Acknowledgement First of all I would like to thank my supervisor Professor Shin Yokoyama for giving
me a great opportunity to work with his laboratory. Without his suggestions and
mentorship, the work would not have been complete ever. I appreciate his dedication
during my entire PhD journey.
I am grateful to thank Professor Takamaro Kikkawa, Professor Masakazu Iwasaka,
Professor Hans Juergn Mattausch, Associate Professor Anri Nakajima, Associate
Professor Shin-Ichiro Kuroki, Associate Professor Tetsushi Koide and Associate
Professor Tetsuo Tabei for their valuable suggestions as well as comments during my
research presentation in the institute meeting.
I am thankful to Professor Akio Kuroda and Assistant Professor Takeshi Ikeda in the
department of Molecular Biotechnology, Graduate School of Advanced Sciences of
Matter (AdSM) for providing prostate specific antigen (PSA) biomarker and valuable
discussion regarding this biomaterial in details.
I am very much grateful to Assistant Professor Yoshiteru Amemiya, in the Research
Institute for Nanodevice and Bio Systems. Without his help I could not have completed
my device fabrication successfully.
I would like to thank the former and current members of Professor Shin Yokoyama
laboratory who helped me a lot either in my clean room or in the device characterization.
I would like to acknowledge financial support from JASSO, special scholarship from
graduate school of Advanced Sciences of Matter (AdSM), research assistantship from
AdSM, teaching assistantship from Research Institute for Nanodevice and Bio Systems
(RNBS).
Finally, I owe my thanks to my wife Lovely Roy and daughter Adrika Aarushi Sana
for being with me at all the good time and bad time as well. My PhD research could not
have finished without their supports and sacrifices.
60
List of publications Journal papers:
(1) Silicon photonic crystal resonators for label free biosensor
A. K. Sana, K. Honzawa, Y. Amemiya, and S. Yokoyama.
Japanese Journal of Applied Physics 55 (2016) 04EM11.
(2) Temperature dependence of resonance characteristics of Silicon resonators and
thermal-stability improvement by differential operation method
A. K. Sana, M. Jun, Shu. Yokoyama, Y. Amemiya, and S. Yokoyama.
Japanese Journal of Applied Physics 56 (2017) 04CC06.
(3) High sensitivity double-cavity silicon photonic-crystal resonator for label free
biosensing
A. K. Sana, Y. Amemiya, and S. Yokoyama.
Japanese Journal of Applied Physics 56 (2017) 04CM06.
61
Conference proceedings
(1) A. K. Sana, Ryuichi Furutani, Yoshiteru Amemiya, and Shin Yokoyama, “Mach-
Zehnder Interferometer Optical Modulator with Cascaded P/N Junctions,” Extended
Abstracts of the 2013 International Conference on Solid State Devices and Materials,
Fukuoka, 2013, pp.1004-1005.
(2) A. K. Sana,Y. Amemiya, and S.Yokoyama, “High Sensitive Biosensor using Si
Photonic Crystal Cavity Resonators,” Extended Abstracts of the 2015 International
Conference on Solid State Devices and Materials, Sapporo, 2015, pp.380-812.
(3) A. K. Sana, Shu. Yokoyama, Y. Nakashima, Y. Amemiya, and Shin.Yokoyama,
“Thermal Change in Resonance Wavelength of Si Resonator Sensors on Si on
Insulator Substrate and Solution by Differential Operation,” Extended Abstracts of
the 2016 International Conference on Solid State Devices and Materials, Tsukuba,
2016, pp.487-488.
(4) A. K. Sana, Y. Amemiya, and S.Yokoyama, “High Sensitivity and High Quality-
factor Silicon Photonic Crystal Resonator with Double Nanocavities for Label free
Bio Sensing,” Extended Abstracts of the 2016 International Conference on Solid
State Devices and Materials, Tsukuba, 2016, pp.361-362.
(5) A. K. Sana, Y. Amemiya, T. Ikeda, A. Kuroda and S.Yokoyama, “Detection of
Prostate Specific Antigen Using Silicon Photonic Crystal Nanocavity Resonator”
Proc. of SPIE Vol. 10111 1011138(1-7), San Francisco 2017 Jan. 28-Feb. 3.
公表論文 (Articles)
(1) Silicon Photonic Crystal Resonators for Label Free Biosensor A. K. Sana, K. Honzawa, Y. Amemiya and S. Yokoyama Japanese Journal of Applied Physics 55 04EM11 (1-5) (2016).
(2) Temperature Dependence of Resonance Characteristics of Silicon Resonators
and Thermal-Stability Improvement by Differential Operation Method A. K. Sana, J. Maeda, Shu. Yokoyama, Y. Amemiya and S. Yokoyama Japanese Journal of Applied Physics 56 04CC06 (1-5) (2017).
(3) High Sensitivity Double-Cavity Silicon Photonic-Crystal Resonator for
Label-Free Biosensing A. K. Sana, Y. Amemiya and S. Yokoyama Japanese Journal of Applied Physics 56 04CM06 (1-5) (2017).